How Significant Are the Real Performance Gains? An Unbiased Evaluation Framework for GraphRAG

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Zeng, Qiming, Yan, Xiao, Luo, Hao, Lin, Yuhao, Wang, Yuxiang, Fu, Fangcheng, Du, Bo, Xu, Quanqing, Jiang, Jiawei
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866918048013746176
author Zeng, Qiming
Yan, Xiao
Luo, Hao
Lin, Yuhao
Wang, Yuxiang
Fu, Fangcheng
Du, Bo
Xu, Quanqing
Jiang, Jiawei
author_facet Zeng, Qiming
Yan, Xiao
Luo, Hao
Lin, Yuhao
Wang, Yuxiang
Fu, Fangcheng
Du, Bo
Xu, Quanqing
Jiang, Jiawei
contents By retrieving contexts from knowledge graphs, graph-based retrieval-augmented generation (GraphRAG) enhances large language models (LLMs) to generate quality answers for user questions. Many GraphRAG methods have been proposed and reported inspiring performance in answer quality. However, we observe that the current answer evaluation framework for GraphRAG has two critical flaws, i.e., unrelated questions and evaluation biases, which may lead to biased or even wrong conclusions on performance. To tackle the two flaws, we propose an unbiased evaluation framework that uses graph-text-grounded question generation to produce questions that are more related to the underlying dataset and an unbiased evaluation procedure to eliminate the biases in LLM-based answer assessment. We apply our unbiased framework to evaluate 3 representative GraphRAG methods and find that their performance gains are much more moderate than reported previously. Although our evaluation framework may still have flaws, it calls for scientific evaluations to lay solid foundations for GraphRAG research.
format Preprint
id arxiv_https___arxiv_org_abs_2506_06331
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle How Significant Are the Real Performance Gains? An Unbiased Evaluation Framework for GraphRAG
Zeng, Qiming
Yan, Xiao
Luo, Hao
Lin, Yuhao
Wang, Yuxiang
Fu, Fangcheng
Du, Bo
Xu, Quanqing
Jiang, Jiawei
Computation and Language
Artificial Intelligence
Information Retrieval
By retrieving contexts from knowledge graphs, graph-based retrieval-augmented generation (GraphRAG) enhances large language models (LLMs) to generate quality answers for user questions. Many GraphRAG methods have been proposed and reported inspiring performance in answer quality. However, we observe that the current answer evaluation framework for GraphRAG has two critical flaws, i.e., unrelated questions and evaluation biases, which may lead to biased or even wrong conclusions on performance. To tackle the two flaws, we propose an unbiased evaluation framework that uses graph-text-grounded question generation to produce questions that are more related to the underlying dataset and an unbiased evaluation procedure to eliminate the biases in LLM-based answer assessment. We apply our unbiased framework to evaluate 3 representative GraphRAG methods and find that their performance gains are much more moderate than reported previously. Although our evaluation framework may still have flaws, it calls for scientific evaluations to lay solid foundations for GraphRAG research.
title How Significant Are the Real Performance Gains? An Unbiased Evaluation Framework for GraphRAG
topic Computation and Language
Artificial Intelligence
Information Retrieval
url https://arxiv.org/abs/2506.06331